US-12619839-B2 - Locating method and locating system for smart shopping cart, computer device and storage medium

US12619839B2US 12619839 B2US12619839 B2US 12619839B2US-12619839-B2

Abstract

The present disclosure provides a locating method and a locating system for a smart shopping cart, a computer device and a storage medium. The method includes: constructing a locating fingerprint database of an smart shopping cart based on commodity attributes collected by an intelligent device and shelf label heartbeat signals received by a communication module; when a motion sensor obtains the motion data of the shopping cart body, the communication module sends all shelf label heartbeat signals received within a second preset time window to a server; the server matches all the shelf label heartbeat signals received within the second preset time window with the locating fingerprint database to obtain locating information of the smart shopping cart.

Inventors

Min Liang
Zhe Fu
Xingyu Zhang

Assignees

Hanshow Technology Co., Ltd.

Dates

Publication Date: 20260505
Application Date: 20250211
Priority Date: 20231113

Claims (20)

1 . A locating method for a smart shopping cart, which is applied to a smart shopping cart in a store having an electronic shelf label system, wherein the electronic shelf label system comprises a server and an electronic shelf label which has known position information and sends a heartbeat data package; the smart shopping cart comprises a shopping cart body, a communication module provided on the shopping cart body and configured to receive the heartbeat data package, an intelligent device provided on the shopping cart body and configured to collect a commodity attribute, the communication module comprises a plurality of antennas and the heartbeat data package comprises a shelf label heartbeat signal; and the method comprises: sending, when the intelligent device obtains the commodity attribute of any commodity, all the shelf label heartbeat signals received by the communication module within the first preset time window, an azimuth at which each of the shelf label heartbeat signals is received, and the commodity attribute to the server in the electronic shelf label system; acquiring, by the server, position information corresponding to the commodity based on the commodity attribute, and obtaining position information corresponding to the shelf label based on shelf label IDs in the shelf label heartbeat signals; storing the position information corresponding to the shelf label and the azimuth in an input data set, and storing the position information corresponding to the commodity in an output data set; and training a machine learning algorithm with the input data set and the output data set to obtain a shopping cart locating model, to process shelf label position information and the azimuth both of which correspond to the shelf label heartbeat signal received using the shopping cart locating model to obtain the locating information of the smart shopping cart.
2 . The locating method for a smart shopping cart according to claim 1 , wherein after obtaining the shopping cart locating model, the method further comprises: inputting shelf label position information and the azimuth both of which correspond to the shelf label heartbeat signal currently received by the communication module into the shopping cart locating model for identification, to obtain current locating information of the smart shopping cart.
3 . The locating method for a smart shopping cart according to claim 1 , wherein the heartbeat data package further comprises a preset sequence of signals, and the method of calculating the azimuth further comprises: calculating the azimuth of the heartbeat signal of each electronic shelf label based on baseband signal characteristics of the preset sequence of signals receive by each antenna.
4 . The locating method for a smart shopping cart according to claim 1 , wherein the heartbeat data package further comprises a preset sequence of signals, and the method further comprises: calculating the azimuth of the heartbeat signal of each electronic shelf label based on baseband signal characteristics of the preset sequence of signals receive by each antenna; and calculating the locating information of the smart shopping cart based on the position information of at least three non-collinear electronic shelf labels and the azimuths corresponding to the at least three non-collinear electronic shelf labels.
5 . The locating method for a smart shopping cart according to claim 4 , wherein the calculating the locating information of the smart shopping cart based on the position information of at least three non-collinear electronic shelf labels and the azimuths corresponding to the at least three non-collinear electronic shelf labels comprises: { ( X 1 - X ) 2 + ( Y 1 - Y ) 2 =  R 1 →  2 ( X 2 - X ) 2 + ( Y 2 - Y ) 2 =  R 2 →  2 ( X 3 - X ) 2 + ( Y 3 - Y ) 2 =  R 3 →  2 {  D 1 →  2 =  R 1 →  2 +  R 2 →  2 + 2 ×  R 1 →  ×  R 2 →  × cos ⁡ ( α 1 - α 2 )  D 2 →  2 =  R 2 →  2 +  R 3 →  2 + 2 ×  R 2 →  ×  R 3 →  × cos ⁢ ( α 2 - α 3 )  D 3 →  2 =  R 3 →  2 +  R 1 →  2 + 2 ×  R 3 →  ×  R 1 →  × cos ⁢ ( α 3 - α 1 ) where (X, Y) represents locating coordinates of the shopping cart, (X 1 , Y 1 ) represents position coordinates of a first electronic shelf label, (X 2 , Y 2 ) represents position coordinates of a second electronic shelf label, (X 3 , Y 3 ) represents position coordinates of a third electronic shelf label, α 1 represents an azimuth of a heartbeat signal of the first electronic shelf label, α 2 represents an azimuth of a heartbeat signal of the second electronic shelf label, α 3 represents an azimuth of a heartbeat signal of the third electronic shelf label, {right arrow over (R 1 )} represents a direction vector from the first electronic shelf label to the shopping cart, {right arrow over (R 2 )} represents a direction vector from the second electronic shelf label to the shopping cart, {right arrow over (R 3 )} represents a direction vector from the third electronic shelf label to the shopping cart, {right arrow over (D 1 )} represents a direction vector from the first electronic shelf label to the second electronic shelf label, {right arrow over (D 2 )} represents a direction vector from the second electronic shelf label to the third electronic shelf label, {right arrow over (D 3 )} represents a direction vector from the third electronic shelf label to the first electronic shelf label.
6 . The locating method for a smart shopping cart according to claim 1 , wherein the smart shopping cart further comprises a motion sensor which is provided on the shopping cart body and configured to collect motion data, and before obtaining the shopping cart locating model, the method further comprises: acquiring, by the server, a shelf serial number and a shelf section index that correspond to each electronic shelf label based on shelf label IDs in all the shelf label heartbeat signals received within the second preset time window, wherein the second preset time window=T4, where T4 is a duration before the motion sensor obtains the motion data; performing aggregation statistics on the shelf label heartbeat signals on the same shelf section based on the shelf serial number and the shelf section index, to obtain a statistical indicator corresponding to each shelf section; and obtaining a target shelf section based on a comprehensive analysis of the statistical indicator corresponding to each shelf section, and taking a coordinate position of the target shelf section as the locating information of the smart shopping cart.
7 . The locating method for a smart shopping cart according to claim 6 , wherein the first preset time window=T1+T2+T3, where T1 is a duration before the intelligent device obtains the commodity attribute, T2 is a duration after the intelligent device obtains the commodity attribute and before the shopping cart starts to move, and T3 is a duration after the shopping cart starts to move.
8 . The locating method for a smart shopping cart according to claim 1 , wherein the acquiring, by the server, target position information of the commodity based on the commodity attribute comprises: if the server acquires a plurality of position information based on the commodity attribute, filtering the plurality of position information based on the locating information of the shopping cart at a previous moment to obtain the target position information of the commodity.
9 . The locating method for a smart shopping cart according to claim 1 , wherein the method further comprises: sending, when the intelligent device obtains a commodity attribute of any commodity, all heartbeat data packages received by the communication module within a first preset time window, an RSSI value of each of the heartbeat data packages received and the commodity attribute to the server in the electronic shelf label system, wherein each of the heartbeat data packages comprises a shelf label heartbeat signal; acquiring, by the server, target position information of the commodity based on the commodity attribute, and binding, by the server, the target position information of the commodity with all the shelf label heartbeat signals received by the communication module within the first preset time window and the RSSI value of each of the heartbeat data packages received, to serve as fingerprint data corresponding to the target position information; and constructing a locating fingerprint database corresponding to a store by acquiring a preset proportion of fingerprint data in the store, to match the shelf label heartbeat signal received on the fingerprint database to obtain the locating information of the smart shopping cart.
10 . The locating method for a smart shopping cart according to claim 9 , wherein the smart shopping cart further comprises a motion sensor which is provided on the shopping cart body and configured to collect motion data, and after constructing a locating fingerprint database corresponding to a store, the method further comprises: sending, by the communication module when the motion sensor obtains motion data of the shopping cart body, all the shelf label heartbeat signals received within a second preset time window to the server, wherein the second preset time window=T4, where T4 is a duration before the motion sensor obtains the motion data; and matching, by the server, all the shelf label heartbeat signals received within the second preset time window with the locating fingerprint database to obtain locating information of the smart shopping cart.
11 . The locating method for a smart shopping cart according to claim 9 , wherein the shelf label heartbeat signal comprises a shelf label ID and a report time; and the fingerprint data corresponding to the target position information comprises a plurality of shelf label IDs and an RSSI weighted average corresponding to each shelf label ID.
12 . The locating method for a smart shopping cart according to claim 9 , wherein after constructing a locating fingerprint database corresponding to a store, the method further comprises: acquiring fingerprint data corresponding to a position of the commodity when the intelligent device continues to obtain the commodity attribute; and updating the locating fingerprint database with the fingerprint data corresponding to the position of the commodity.
13 . A locating system for a smart shopping cart, comprising an electronic shelf label which has known position information, a server and a smart shopping cart, wherein the smart shopping cart comprises a shopping cart body, a communication module provided on the shopping cart body and configured to receive a heartbeat data package, and an intelligent device provided on the shopping cart body and configured to collect a commodity attribute, wherein the communication module comprises a plurality of antennas and the heartbeat data package comprises a shelf label heartbeat signal; the electronic shelf label is configured to send a heartbeat data package; the communication module is configured to, when the intelligent device obtains the commodity attribute of any commodity, send all the shelf label heartbeat signals received by the communication module within the first preset time window, an azimuth at which each of the shelf label heartbeat signals is received, and the commodity attribute to the server in the electronic shelf label system; the server is configured to acquire position information corresponding to the commodity based on the commodity attribute, and obtain position information corresponding to the shelf label based on shelf label IDs in the shelf label heartbeat signals; the server is configured to store the position information corresponding to the shelf label and the azimuth in an input data set, and store the position information corresponding to the commodity in an output data set; the server is configured to train a machine learning algorithm with the input data set and the output data set to obtain a shopping cart locating model, to process shelf label position information and the azimuth both of which correspond to the shelf label heartbeat signal received using the shopping cart locating model to obtain the locating information of the smart shopping cart.
14 . The locating system for a smart shopping cart according to claim 13 , wherein the locating system further comprises: the server is configured to input shelf label position information and the azimuth both of which correspond to the shelf label heartbeat signal currently receive by the communication module into the shopping cart locating model for identification, to obtain current locating information of the smart shopping cart.
15 . The locating system for a smart shopping cart according to claim 13 , wherein the heartbeat data package further comprises a preset sequence of signals, and the locating system further comprises: the communication module further configured to calculate the azimuth of heartbeat signal of each electronic shelf label based on baseband signal characteristics of the preset sequence of signals received by each antenna.
16 . The locating system for a smart shopping cart according to claim 13 , wherein the heartbeat data package further comprises a preset sequence of signals, and the locating system further comprises: the communication module further configured to calculate the azimuth of heartbeat signal of each electronic shelf label based on baseband signal characteristics of the preset sequence of signals received by each antenna; and the server further configured to calculate the locating information of the smart shopping cart based on the position information of at least three non-collinear electronic shelf labels and the azimuths corresponding to the at least three non-collinear electronic shelf labels.
17 . The locating system for a smart shopping cart according to claim 13 , wherein the server further configured to: acquire a shelf serial number and a shelf section index that correspond to each electronic shelf label based on shelf label IDs in all the shelf label heartbeat signals received within the second preset time window, wherein the second preset time window=T4, where T4 is a duration before the motion sensor obtains the motion data; perform aggregation statistics on the shelf label heartbeat signals on the same shelf section based on the shelf serial number and the shelf section index, to obtain a statistical indicator corresponding to each shelf section; and obtain a target shelf section based on a comprehensive analysis of the statistical indicator corresponding to each shelf section, and taking a coordinate position of the target shelf section as the locating information of the smart shopping cart.
18 . The locating system for a smart shopping cart according to claim 13 , wherein the locating system further comprises: the communication module is further configured to send, when the intelligent device obtains a commodity attribute of any commodity, all the heartbeat data packages received within a first preset time window and the commodity attribute to the server in an electronic shelf label system; the server is further configured to acquire target position information of the commodity based on the commodity attribute, and bind the target position information of the commodity with all the shelf label heartbeat signals received by the communication module within the first preset time window, to serve as fingerprint data corresponding to the target position information, and construct a locating fingerprint database corresponding to a store by acquiring a preset proportion of fingerprint data in the store, to match the shelf label heartbeat signal received on the fingerprint database to obtain the locating information of the smart shopping cart.
19 . The locating system for a smart shopping cart according to claim 18 , wherein the smart shopping cart further comprises a motion sensor which is provided on the shopping cart body and configured to collect motion data, and the locating system further comprises: the communication module is further configured to send, when the motion sensor obtains the motion data of the shopping cart body, all the shelf label heartbeat signals received within a second preset time window to the server, wherein the second preset time window=T4, where T4 is a duration before the motion sensor obtains the motion data; and the server is further configured to match all the shelf label heartbeat signals received within the second preset time window with the locating fingerprint database to obtain locating information of the smart shopping cart.
20 . A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when executing the computer program, the processor implements the steps of a method, the method is applied to a smart shopping cart in a store having an electronic shelf label system, wherein the electronic shelf label system comprises a server and an electronic shelf label which has known position information and sends a heartbeat data package; the smart shopping cart comprises a shopping cart body, a communication module provided on the shopping cart body and configured to receive the heartbeat data package, an intelligent device provided on the shopping cart body and configured to collect a commodity attribute, the communication module comprises a plurality of antennas and the heartbeat data package comprises a shelf label heartbeat signal; and the method comprises: sending, when the intelligent device obtains the commodity attribute of any commodity, all the shelf label heartbeat signals received by the communication module within the first preset time window, an azimuth at which each of the shelf label heartbeat signals is received, and the commodity attribute to the server in the electronic shelf label system; acquiring, by the server, position information corresponding to the commodity based on the commodity attribute, and obtaining position information corresponding to the shelf label based on shelf label IDs in the shelf label heartbeat signals; storing the position information corresponding to the shelf label and the azimuth in an input data set, and storing the position information corresponding to the commodity in an output data set; and training a machine learning algorithm with the input data set and the output data set to obtain a shopping cart locating model, to process shelf label position information and the azimuth both of which correspond to the shelf label heartbeat signal received using the shopping cart locating model to obtain the locating information of the smart shopping cart.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of U.S. patent application Ser. No. 18/825,629, filed on Sep. 5, 2024, which claims priority to Chinese Patent Application No. 202311505182.3, filed on Nov. 13, 2023, both of which are hereby incorporated by reference their entireties. TECHNICAL FIELD The present disclosure relates to the field of locating technology, and particularly to a locating method and a locating system for smart shopping cart, a computer device and a storage medium. BACKGROUND At present, large supermarkets around the world are carrying out digital upgrading and transformation. In the digital applications, electronic shelf labels will replace the traditional paper labels. In addition to displaying ordinary information, electronic shelf labels can also be used in many applications, such as quick picking, out-of-stock management, quick inventory and man-machine interaction for users. Regarding a shopping cart walking in a shelf area, it is usually necessary to locate the shopping cart to provide a navigation thereof and a service recommendation based on the position of the shopping cart. At present, the Bluetooth locating technology is usually used for the locating of the shopping cart. In order to cover all the shelf aisles in large supermarkets, a large number of Bluetooth beacons need to be deployed, which leads to the problems of complicated installation and high maintenance cost. SUMMARY Aiming at the disadvantages in the prior art, the present disclosure provides a locating method and a locating system for a smart shopping cart, a computer device and a storage medium, which solve the problems of complicated installation and high maintenance cost in the shopping cart locating method in the prior art. The present disclosure realizes a fingerprint locating method for the shopping cart based on an electronic shelf label system applied in a store, thereby reducing the hardware installation complexity of the locating technology and reducing maintenance cost thereof, and improving the locating accuracy and stability. In a first aspect, the present disclosure provides a locating method for a smart shopping cart, which is applied to a smart shopping cart in a store having an electronic shelf label system, and the electronic shelf label system includes a server and an electronic shelf label which has known position information and sends heartbeat data package; the smart shopping cart includes a shopping cart body, and a communication module provided on the shopping cart body and configured to receive the heartbeat data package, a motion sensor provided on the shopping cart body and configured to collect motion data, and an intelligent device provided on the shopping cart body and configured to collect a commodity attribute, the method includes: sending, when the intelligent device obtains a commodity attribute of any commodity, all heartbeat data packages received by the communication module within a first preset time window, an RSSI value of each of the heartbeat data packages received and the commodity attribute to the server in the electronic shelf label system, and each of the heartbeat data packages includes a shelf label heartbeat signal; acquiring, by the server, target position information of the commodity based on the commodity attribute, and binding the target position information of the commodity with all the shelf label heartbeat signals received by the communication module within the first preset time window and the RSSI value of each of the heartbeat data packages received, to serve as fingerprint data corresponding to the target position information; and constructing a locating fingerprint database corresponding to a store by acquiring a preset proportion of fingerprint data in the store; sending, by the communication module; sending, by the communication module when the motion sensor obtains motion data of the shopping cart body, all the shelf label heartbeat signals received within a second preset time window to the server; and matching, by the server, all the shelf label heartbeat signals received within the second preset time window with the locating fingerprint database to obtain locating information of the smart shopping cart. Optionally, the shelf label heartbeat signal includes a shelf label ID and a report time; the fingerprint data corresponding to the target position information includes a plurality of shelf label IDs and an RSSI weighted average corresponding to each shelf label ID. Optionally, the first preset time window=T1+T2+T3, where T1 is a duration before the intelligent device obtains the commodity attribute, T2 is a duration after the intelligent device obtains the commodity attribute and before the shopping cart starts to move, and T3 is a duration after the shopping cart starts to move; and the second preset time window T4, where T4 is a duration before the motion sensor obtains the motion data. Optionally, the acquiring, by th